© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 447
    Prev
    Next
    INDEX
    Explanations

    electroconvulsive therapy

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron specifically detects mentions of “electro‐” treatments or terms—most strongly electroconvulsive therapy (ECT) and related “electro” prefixes.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    árna
    -0.71
     sneezing
    -0.65
     calyx
    -0.65
    romechanical
    -0.65
     barometer
    -0.64
    ἧ
    -0.63
     Catalytic
    -0.63
     Benedetto
    -0.63
    atima
    -0.63
    惯
    -0.63
    POSITIVE LOGITS
     electric
    1.85
     shock
    1.74
     electro
    1.65
     electricity
    1.57
     shocks
    1.55
     Electric
    1.55
    Electric
    1.55
     Electro
    1.54
    Electro
    1.54
     electrical
    1.53
    Activations Density 0.070%

    No Known Activations